Use of GP data identifies a third more patients with atrial fibrillation than hospital data alone, a team at the University of Oxford have found.
Researchers trawled through patient records for 230,000 participants taking part in UK Biobank to see potential differences between people with atrial fibrillation recorded in different healthcare settings.
Over a seven-year follow up period 7,136 had a diagnosis of atrial fibrillation recorded.
The study identified 28% more cases of atrial fibrillation in primary care data than secondary care data alone.
In addition, evidence of atrial fibrillation was also recorded in GP records an average of 1.3 years earlier, they reported in the Europace journal.
Lower rates of anticoagulation therapy and higher rates of death were seen in patients with atrial fibrillation that had only been recorded in hospital compared to those recorded in primary care.
And atrial fibrillation only recorded in hospital data was associated with more established heart disease and prior medication use than that recorded in primary care.
From a practical point of view, it highlights the importance of linking up primary and secondary care records as well as timely communication across NHS settings, the researchers said.
Their findings also support the Sudlow Review published last year which found that establishing a central system to allow access to GP data for research should be of the highest priority.
Professor Jemma Hopewell, professor of precision medicine and epidemiology at Oxford Population Health and senior author of the study, said: ‘Access to electronic healthcare records has revolutionised large-scale epidemiological research, but we need to consider the richness of the data available and not rely on hospital records alone to study chronic conditions such as atrial fibrillation.
‘Our findings show that data from across healthcare settings are extremely powerful for understanding patient profiles, management and consequences of atrial fibrillation.
‘These insights can then be used to develop new strategies that will allow us to improve patient outcomes.’
She added that using hospital records alone may not be enough to accurately assess the relevance of different risk factors, understand disease progression, or look at the effectiveness of implementation strategies for managing the condition.
Professor Bryan Williams, chief scientific and medical officer at the British Heart Foundation, said it was crucial that atrial fibrillation was detected at an early stage ‘so that doctors can take steps to treat the condition and prevent more strokes’.
He added: ‘This study addresses a hugely important challenge for the modern era of wearable technologies, notably, how to integrate data from technology into NHS data systems so that clinicians can actually access and act on the data.
‘The study shows that when this integration of the patient’s primary care and hospital records is achieved, many more cases of atrial fibrillation can be identified, allowing doctors to act to limit its consequences.
‘Helping researchers and clinicians to do this, while maintaining people’s trust in the use of this data, must be a focus for the Government.’
Pulse October survey
Take our April 2025 survey to potentially win £200 worth of tokens
Back in the day of mercury spygs. beat to beat variation was a quick clue to AF. When yo were listening in it was always notable when you picked up an variable loss of beats for the diastolic. You don’t get that with electronic sphygs.